Thermo-mechanical performance evaluation of hybrid NiTi/CF-PEKK composite laminates using experiments and machine learning approaches

Background: Suicide remains a leading cause of death among youth, yet effective tools to predict suicide attempts (SA) in individuals under 18 are scarce. This study aims to develop machine learning (ML) models to predict SA in paediatric populations using Google Trends data. Methods: Relative Search Volumes (RSVs) from Google Trends were analysed for terms linked to suicide risk factors. Pearson Correlation Coefficients (PCC) identified terms strongly associated with SA rates. Based on these, several ML models were developed and evaluated, including Random Forest Regression, Support Vector Regression (SVR), XGBoost, and Linear Regression. Model performance was assessed using metrics such as PCC, mean absolute error (MAE), mean squared error (MSE), root mean square error (RMSE), and mean absolute percentage error (MAPE). Results: Terms related to suicide prevention and symptoms, including psychiatrist and anxiety disorder, showed the strongest correlations with SA rates (PCC ≥ 0.90). Random Forest Regression emerged as the top-performing ML model (PCC = 0.953, MAPE = 20.12%, RMSE = 17.21), highlighting burnout, anxiety disorder, antidepressants, and psychiatrist as key predictors of SA. Other models’ scores were XGBoost (PCC = 0.446, MAPE = 22.57%, RMSE = 18.03), SVR (PCC = 0.833, MAPE = 42.23%, RMSE = 47.32) and Linear Regression (PCC = 0.947, MAPE = 23.64%, RMSE = 17.66). Conclusions: Google Trends–based ML models suggest potential utility for short-term prediction of youth SA. These preliminary findings support the utility of search data in identifying real-time suicide risk in paediatric populations.

Predicting egg production rate and egg weight of broiler breeders based on machine learning and Shapley additive explanations

The permeability coefficient of soils is an essential measure for designing geotechnical construction. The aim of this paper was to select a highest performance and reliable machine learning (ML) model to predict the permeability coefficient of soil and quantify the feature importance on the predicted value of the soil permeability coefficient with aided machine learning‐based SHapley Additive exPlanations (SHAP) and Partial Dependence Plot 1D (PDP 1D). To acquire this purpose, five single ML algorithms including K‐nearest neighbors (KNN), support vector machine (SVM), light gradient boosting machine (LightGBM), random forest (RF), and gradient boosting (GB) are used to build ML models for predicting the permeability coefficient of soils. Performance criteria for ML models include the coefficient of correlation R 2 , root mean square error (RMSE), mean absolute percentage error (MAPE), and mean absolute error (MAE). The best performance and reliable single ML model for predicting the permeability coefficient of soil for the testing dataset is the gradient boosting (GB) model, which has R 2 = 0.971, RMSE = 0.199 × 10 −11 m/s, MAE = 0.161 × 10 −11 m/s, and MAPE = 0.185%. To identify and quantify the feature importance on the permeability coefficient of soil, sensitivity studies using permutation importance, SHapley Additive exPlanations (SHAP), and Partial Dependence Plot 1D (PDP 1D) are performed with the aided best performance and reliable ML model GB. Plasticity index, density > water content, liquid limit, and plastic limit > clay content > void ratio are the order effects on the predicted value of the permeability coefficient. The plasticity index and density of soil are the first priority soil properties to measure when assessing the permeability coefficient of soil.

Optimized machine learning models for predicting crown convergence of plateau mountain tunnels

<p>Accurate water level (WL) forecasting is important for water resources management and planning purposes in the Great Lakes. The objectives of this research are two-fold.  The first objective is to apply machine learning (ML) (i.e., random forest (RF) and support vector regression (SVR)) and hybrid convolutional neural network(CNN)-long-short term memory (LSTM) deep learning (DL) models for multi-step (i.e., one-, two- and three-monthly step ahead) WL forecasting in the Great Lakes (Michigan and Ontario). The second objective is to integrate the boundary corrected (BC) maximal overlap discrete wavelet transform (MODWT) with SVR, RF, and CNN-LSTM models to improve the performance of the individual models. By employing a BC-wavelet decomposition method, the ‘future data’ issue (i.e., data from the future that is not available), often overlooked in the literature and a major barrier to achieving realistic forecasting performance is overcome. </p><p>For Lakes Michigan and Ontario, 1212 monthly WL (m) records (spanning Jan 1918–Dec 2018) were used to develop the models. For the non-wavelet-based models (SVR, RF, and CNN-LSTM), candidate model inputs included the WL recorded over the previous 12 months.  For the BC-MODWT-based models (BC-MODWT-SVR, BC-MODWT-RF, and BC-MODWT-CNN-LSTM), the lagged input time series were decomposed into BC-wavelet and scaling coefficients by using different mother wavelets (Haar, Daubechies, Symlets, Fejer-Korovkin and Coiflets), filter lengths (from two up to 12) and decomposition levels (from one up to seven).  For each method (SVR, RF, and CNN-LSTM), mother wavelet, and decomposition level a model was generated.  For both wavelet- and non-wavelet-based models, the particle swarm optimization (PSO) method was used to select the most appropriate inputs to include in the proposed multi-step WL forecasting models.</p><p>The datasets were partitioned into calibration and validation subsets. After calibrating the models, various performance evaluation metrics, e.g., coefficient of determination (R<sup>2</sup>), root mean square error (RMSE), mean absolute error (MAE), root mean square percentage error (RMSPE), mean absolute percentage error (MAPE) and the Nash-Sutcliffe efficiency coefficient (NSC) were used to assess model accuracy.</p><p>Of the ML models, the SVR outperformed RF while the DL models outperformed the ML models for each forecast lead time (one-, two-, and three-step(s) ahead). Results from this case study indicate that not all wavelet families and decomposition levels perform equally and in some cases, the wavelet-based models do not improve performance over the non-wavelet-based models. However, the BC-MODWT-CNN-LSTM using suitable mother wavelets (e.g., Haar) outperforms the individual ML and BC-MODWT-ML-based models. More accurate forecasts were obtained for Lake Michigan although the performance in both Great Lakes was accurate. The outcomes of this research indicate that the BC-MODWT-CNN-LSTM model is a promising tool for generating accurate WL forecasts.</p>

Prediction of deflections of reinforced concrete (RC) flexural structures is vital to evaluate the workability and safety of structures during its life cycle. Empirical methods are limited to predict a long-term deflection of RC structures because they are difficult to consider all influencing factors. This study presents data-driven machine learning (ML) models to early predict the long-term deflections in RC structures. An experimental dataset was used to build and evaluate single and ensemble ML models. The models were trained and tested using the stratified 10-fold cross-validation algorithm. Analytical results revealed that the ML model is effective in predicting the deflection of RC structures with good accuracy of 0.972 in correlation coefficient (R), 8.190 mm in root mean square error (RMSE), 4.597 mm in mean absolute error (MAE), and 16.749% in mean absolute percentage error (MAPE). In performance comparison against with empirical methods, the prediction accuracy of the ML model improved significantly up to 66.41% in the RMSE and up to 82.04% in the MAE. As a contribution, this study proposed the effective ML model to facilitate designers in early forecasting long-term deflections in RC structures and evaluating their long-term serviceability and safety.

Improving carbon flux estimation in tea plantation ecosystems: A machine learning ensemble approach

Due to many substances in the human exposome, there is a dearth of exposure and toxicity information available to assess potential health risks. Quantification of all trace organics in the biological fluids seems impossible and costly, regardless of the high individual exposure variability. We hypothesized that the blood concentration () of organic pollutants could be predicted via their exposure and chemical properties. Developing a prediction model on the annotation of chemicals in human blood can provide new insight into the distribution and extent of exposures to a wide range of chemicals in humans. Our objective was to develop a machine learning (ML) model to predict blood concentrations () of chemicals and prioritize chemicals of health concern. We curated the of compounds mostly measured at population levels and developed an ML model for chemical predictions by considering chemical daily exposure (DE) and exposure pathway indicators (), half-lives (), and volume of distribution (). Three ML models, including random forest (RF), artificial neural network (ANN) and support vector regression (SVR) were compared. The toxicity potential or prioritization of each chemical was represented as a bioanalytical equivalency (BEQ) and its percentage (BEQ%) estimated based on the predicted and ToxCast bioactivity data. We also retrieved the top 25 most active chemicals in each assay to further observe changes in the BEQ% after the exclusion of the drugs and endogenous substances. We curated the of 216 compounds primarily measured at population levels. RF outperformed the ANN and SVF models with the root mean square error (RMSE) of 1.66 and , the mean absolute error (MAE) values of 1.28 and , the mean absolute percentage error (MAPE) of 0.29 and 0.23, and of 0.80 and 0.72 across test and testing sets. Subsequently, the human of 7,858 ToxCast chemicals were successfully predicted, ranging from to . The predicted were then combined with ToxCast in vitro bioassays to prioritize the ToxCast chemicals across 12 in vitro assays with important toxicological end points. It is interesting that we found the most active compounds to be food additives and pesticides rather than widely monitored environmental pollutants. We have shown that the accurate prediction of "internal exposure" from "external exposure" is possible, and this result can be quite useful in the risk prioritization. https://doi.org/10.1289/EHP11305.

The article demonstrates the Brix content of melon fruits grafted with different varieties of rootstock using Support Vector Regression (SVR) and Multiple Linear Regression (MLR) model approaches. The analysis yielded primary fruit biochemical measurements on the following rootstocks, Sphinx, Albatros, and Dinero: nitrogen, phosphorus, potassium, calcium, and magnesium. Established models were evaluated with Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Mean Square Error (MSE), Root Mean Square Error (RMSE), and Coefficient of Determination (R2) metrics. In the test section, the results of the MLR model were calculated as MAE: 0.0728, MAPE: 0.0117, MSE: 0.0088, RMSE: 0.0936, and R2: 0.9472, while the results of the SVR model were calculated as MAE: 0.0334, MAPE: 0.0054, MSE: 0.0016, RMSE: 0.0398, and R2: 0.9904. Despite both models performing well, the SVR model showed superior accuracy, outperforming MLR by 54% to 82% in terms of predictions. The relationships between Brix levels and various nutrients, such as sucrose, glucose, and fructose, were found to be strong, while titratable acidity had a minimal effect. SVR was found to be a more reliable, non-destructive method for melon quality assessment. These findings revealed the relationship between Brix and sugar levels on melon quality. The study highlights the potential of these machine learning models in optimizing the rootstock effect and managing melon cultivation to improve fruit quality.

PurposeThe purpose of this research was to develop and evaluate a machine learning (ML) algorithm to accurately predict bamboo compressive strength (BCS). Using a dataset of 150 bamboo samples with features such as cross-sectional area, dry weight, density, outer diameter, culm thickness and load, various ML algorithms including artificial neural network (ANN), extreme learning machine (ELM) and support vector regression (SVR) were tested. The ELM algorithm outperformed others, showing superior accuracy based on metrics like R2, MSE, RMSE, MAE and MAPE. The study highlights the efficacy of ELM in enhancing the precision and reliability of BCS predictions, establishing it as a valuable tool for assessing bamboo strength.Design/methodology/approachThis study experimentally created a dataset of 150 bamboo samples to predict BCS using ML algorithms. Key predictive features included cross-sectional area, dry weight, density, outer diameter, culm thickness and load. The performance of various ML algorithms, including ANN, ELM and SVR, was evaluated. ELM demonstrated superior performance based on metrics such as coefficient of determination (R2), mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE), establishing its robustness in predicting BCS accurately.FindingsThe study found that the ELM algorithm outperformed other ML algorithms, including ANN and SVR, in predicting BCS. ELM achieved the highest accuracy based on key metrics such as R2, MSE, RMSE, MAE and MAPE. These results indicate that ELM is a highly effective and reliable tool for predicting the compressive strength of bamboo, thereby enhancing the precision and dependability of BCS evaluations.Originality/valueThis study is original in its application of the ELM algorithm to predict BCS using experimentally derived data. By comparing ELM with other ML algorithms like ANN and SVR, the research establishes ELM’s superior performance and reliability. The findings demonstrate the significant potential of ELM in material strength prediction, offering a novel and robust approach to evaluating bamboo’s compressive properties. This contributes valuable insights into the field of material science and engineering, particularly in the context of sustainable construction materials.

Broadband high gain performance MIMO antenna array for 5 G mm-wave applications-based gain prediction using machine learning approach

Microbars are one of the important components of microelectromechanical systems. With the recent increase in their applications, the importance of understanding their mechanical response has become an important topic. In this study, for the first time, the mechanical behavior of microbars based on the strain gradient theory is investigated using a machine learning (ML) approach. Four distinct ML models, namely artificial neural network (ANN), support vector regression (SVR), decision tree regression (DTR), and random forest regression (RFR), are developed for microbars with clamped boundary conditions. The performance of these models is individually assessed using five different metrics: coefficient of determination (R2), Mean Absolute Error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE) and Nash‐Sutcliffe efficiency coefficient (NSE). The best‐performing model is selected based on these comparisons. Additionally, a semi‐analytical approach is employed to determine the natural frequencies of microbars under general elastic boundary conditions using the Fourier sine series and Stokes' transform. While theR2value for all four models indicated a good fit of 0.999, the percentage difference in MAE and RMSE values between the training and testing data for DTR and RFR models was relatively higher as compared to ANN and SVR. The results showed that ANN and SVR models exhibit the best performance in predicting the natural frequencies on both training and testing data across all three metrics. Finally, a study on the free axial vibration frequencies of microbar under various effects was conducted.

Accuracy and uncertainty of models used for digital soil mapping are important for assessing confidence of predictions and reliable land use planning and management. In this study, two approaches of geostatistical (spatial) and machine learning (ML) models were evaluated forpredictivemapping of soil calcium (Ca) and potassium (K). Two spatial models including empirical Bayesian kriging (EBK) and sequential Gaussian simulation (SGS) were compared with machine learning models: Cubist, random forest (RF) and support vector machine (SVM) in terms of their accuracy and uncertainty for mapping soil Ca and K. The study area is in Nowley, New South Wales, Australia, with an area of 2083ha and a variety of soil types and farming systems. For the models training process, 240 soil samples data and for validation 102 independent samples data were used. For accuracy assessment R2, root mean square error (RMSE), concordance and bias and for uncertainty assessment confidence limits were investigated. Also, in order to compare the outcomes for the two soil properties with different measurement units, mean absolute percentage error (MAPE) and relative uncertainty (RU) as accuracy and uncertainty measures, respectively, were evaluated. Results showed that for K map SGS had the highest R2 (0.74) and lowest RMSE (1.96), followed by EBK with R2 = 0.72 and RMSE = 2.02. For Ca map, EBK model showed the highest accuracy (R2 = 0.46; RMSE = 3.21), followed by SVM and SGS with comparable accuracies. Comparing the two soil properties, Ca map showed higher MAPE and RU, compared to K map. The lowest MAPE was obtained for EBK model (for K = 39) and SGS model (for K = 44). Also, the lowest RU values were found for EBK and SGS models. Among the ML models, SVM showed lower sensitivity to higher variance in data input. In general, the spatial models outperformed the ML models with regard to both accuracy and uncertainty. An additional conclusion is that considering the data variance in the two soil properties, geostatistical models with lower RU and MAPE were relatively less susceptible to data variance, compared to ML models.

Application of machine learning in predicting corrosion inhibition capacity of Spinacia oleracea leaf extract on copper

Tomato is one of the most important vegetables due to its high production and nutritional value. With the development of digital agriculture, the tomato breeding and processing industries have seen a rapid increase in the need for simple, low-labor, and inexpensive methods for analyzing tomato composition. This study proposes a digital method to predict four minerals (calcium, potassium, phosphorus, and magnesium) in beef-type tomato using machine learning models, including k-nearest neighbors (kNN), artificial neural networks (ANNs), and Support Vector Regression (SVR). The models were discriminated using the coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). The kNN model showed the best performance for estimation of quantity of calcium, potassium, phosphorus, and magnesium. The results demonstrate that kNN consistently outperforms ANNs and SVR across all target nutrients, achieving the highest R2 and the lowest error metrics (RMSE, MAE, and MAPE). Notably, kNN achieved an exceptional R2 of 0.8723 and a remarkably low MAPE of 3.95% in predicting phosphorus. This study highlights how machine learning can provide a versatile, accurate, and efficient solution for tomato mineral analysis in digital agriculture.